The Agent Skills Directory

[COMMAND_EXECUTION]: The utility script scripts/compute-target-versions.mjs uses execSync to run a static npm view command. This is used to retrieve version information from the npm registry to help the user select the correct target for a migration.
Evidence: Found in scripts/compute-target-versions.mjs: execSync('npm view nx dist-tags --json', ...).- [INDIRECT_PROMPT_INJECTION]: The skill's workflow involves the agent reading external, untrusted data such as upstream changelogs and migration guides to determine which code changes require migrations. This creates a surface where malicious instructions in external documentation could potentially influence the agent's migration authoring process.
Ingestion points: External changelogs, migration guides, and third-party repositories referenced in SKILL.md (Section 1).
Boundary markers: The instructions do not specify explicit delimiters or "ignore instructions" warnings for the external content being analyzed.
Capability inventory: The agent has access to file system tools (Read, Write, Edit, Glob, Grep) and Bash to implement the migrations identified from the data.
Sanitization: No specific sanitization or validation of the external content is mentioned before processing.

author-migration